169results about How to "Easy to capture" patented technology

The invention discloses a power electronic transformer working mode recognition method and device, equipment, a medium and a product, and relates to the field of artificial intelligence, and the method comprises the steps: collecting original inductive current data during the operation of a power electronic transformer; performing adaptive segmentation normalization processing on the original inductive current data to generate a normalized inductive current sequence; calculating a wavelet packet energy entropy and a time domain differential entropy of the normalized inductive current sequence, and splicing the wavelet packet energy entropy and the time domain differential entropy into a two-dimensional fusion feature vector; inputting the normalized inductive current sequence and the two-dimensional fusion feature vector into a trained deep learning model to obtain a prediction probability vector; based on the prediction probability vector, the working mode category with the maximum probability value is selected as the recognition result, and the recognition precision of the working modes of the power electronic transformer can be guaranteed under the working conditions of high noise interference or rapid mode switching.

The embodiment of the present specification provides a task processing model training method, a task processing method and a request processing method based on the task processing model. The task processing model training method comprises the following steps: obtaining a sample set and a test result of a sample task; determining a key processing unit corresponding to the sample task from a plurality of processing units according to the test result, wherein the influence degree of the key processing unit on the test result is greater than the influence degree of a non-key processing unit on the test result, and the non-key processing unit is a processing unit other than the key processing unit in the task processing model; and adjusting parameters of the key processing unit according to the sample set to obtain a trained task processing model. The key processing unit is determined before the model training, and only the parameters of the key processing unit corresponding to the sample task are adjusted, which reduces the mutual interference between different tasks, effectively solves the performance conflict problem, and improves the multi-task processing performance of the task processing model.

The present application relates to computer vision and underwater detection technology, provide a kind of underwater small biological target detection method and system based on deep learning, the WLDA-YOLOv8 deep learning model constructed includes main network, neck network and detection head, main and neck embed light multi-scale module, detection head introduces dynamic detection head and adaptive sample selection strategy;Training stage uses improved WIoUv3s loss to carry out boundary box regression, the model is trained using underwater biological data set, finally realizes the accurate detection of small target in underwater image, enhances the capture and fusion ability to small target subtle features.

The application discloses a space-time convolution short-time traffic flow prediction method based on complementary ensemble empirical mode decomposition, and comprises the following steps: acquiring original traffic flow data and performing data preprocessing; and inputting the traffic flow data into a pre-constructed CEEMD-GAT-LSTM model to perform short-time traffic flow prediction. The prediction model comprises the following steps: performing signal decomposition on the traffic flow data by using CEEMD, fully mining the spatial features of the data by using GAT, and fully mining the time features of the data by using LSTM, so that the learning efficiency of the neural network is improved. The application can effectively improve the prediction accuracy of short-time traffic flow prediction and provide favorable reference value for the traffic department.

This invention discloses a method and system for predicting solar radio bursts based on a complete solar magnetograph, belonging to the field of solar radio technology. It includes: acquiring a complete solar magnetograph; inputting the complete solar magnetograph into a pre-trained improved ResNet model to obtain image feature vectors; dividing the image feature vectors into pseudo-time series and inputting them into an improved LSTM network to obtain solar radio burst prediction results; the improved ResNet model introduces a custom residual layer to extract small-scale features from the complete solar magnetograph, and the improved LSTM network introduces a gating mechanism to process the image feature vectors at different time scales. This method can improve the accuracy of solar radio burst prediction without utilizing solar radio spectrum data; it solves the problem that existing predictions using solar radio spectra are easily interfered with by the accuracy of the input data observations.

The invention discloses a motion intention recognition method based on multi-modal enhanced comparative learning, and the method achieves the precise motion intention recognition based on any single mode through systematic fusion of multi-modal physiological signals and an innovative comparative learning architecture. Intention perception contrast learning: constructing positive and negative sample pairs enhanced by a multi-modal time sequence and optimizing contrast loss, so that the discrimination capability of the model in dynamic intention transfer is remarkably improved; according to many-for-one cross-modal prediction, multi-modal identification performance under single-modal input is realized through semantic alignment and a knowledge migration mechanism. After pre-training joint optimization, training and reasoning can be completed only through single-modal data, and data dependence is remarkably reduced while the advantage of multi-modal performance is kept. According to the technical scheme, the accuracy and robustness in lower limb movement intention recognition in the brain-computer interaction field are effectively improved, and an innovative technical solution is provided for a movement auxiliary rehabilitation system.

The invention relates to an image processing method, a map generation method, electronic equipment and a storage medium. The image processing method comprises the following steps: extracting image features from an environment image to obtain a feature map of the environment image; segmenting the feature map through at least two different types of geometric element models to obtain different types of geometric element maps; fusing the geometric element diagrams of different categories to obtain a first fused diagram; and performing vectorization processing on geometric elements in the first fusion image to obtain a structured vector graph corresponding to the environment image.

The invention belongs to the technical field of health state assessment, and discloses a health state assessment method and device, electronic equipment and a storage medium, and the method comprises the steps: obtaining the physiological feature data of a user in a preset period, constructing the physiological feature data into tensor data containing an individual dimension, a feature dimension and a time dimension, complementing missing data in the tensor data by using a frequency domain transformation method and a singular value decomposition method to obtain complemented physiological feature data, inputting the complemented physiological feature data into a preset health state evaluation model, and calculating to obtain a health state evaluation result of the user in a preset period; by inputting the complemented physiological feature data obtained by complementing with the frequency domain transformation method and the singular value decomposition method into the preset health state assessment model, the health state assessment result of the user in the preset period is calculated, and the accuracy of health state assessment is improved.

The device comprises an ash hopper shell, a flue is arranged on the outer side of the ash hopper shell, a plurality of hanging lug plates are symmetrically arranged on the inner walls of the two sides of the ash hopper shell, and a flow guide plate is arranged between every two corresponding hanging lug plates. According to the invention, the ash bucket structure in which the plurality of hanging lug plates and the guide plates are matched with each other is arranged in the ash bucket shell, so that a flow field can be changed by utilizing the additionally arranged guide plates, and dust in flue gas can be prevented from being directly taken out of the ash bucket shell by bypassing the flow field; moreover, due to the arrangement of the guide plate, the through-flow area of the ash bucket shell is reduced, the flow velocity of the flue gas is increased, the flue gas has a larger turn under the flow field deviation action of the guide plate, and the dust can be thrown to the bottom of the ash bucket shell to be deposited under the action of centrifugal force and gravity, so that the capture and the sedimentation of the dust are facilitated. And meanwhile, the contact area between the flue gas and the ash bucket shell is increased due to the arrangement of the guide plate, so that dust is better captured.

The utility model discloses an electric decoy duck propeller, including propeller subassembly, timer component, power connection line and battery output line, wherein propeller subassembly connects duck decoy, is placed in water and provides downward propulsion power, controls duck decoy and makes the action of diving in water in water, the output end of power connection line is connected to propeller subassembly, and propeller subassembly is provided with the transmission of underwater signal and electric energy, the output end of timer component is connected to the input end of power connection line, and propeller subassembly is provided with control signal, and the frequency and length of time of duck decoy diving are controlled, and battery output line is connected to the input end of timer component, and propeller subassembly is provided with electric energy. The utility model discloses simple structure, low in production cost, easy operation, high practicality can realize the action of the multiple electric duck to underwater movement to imitate duck natural eating to reach the effect of attracting duck landing water surface, so that better hunting.

The application discloses a composite solid decoloring powder and a preparation method thereof, and belongs to the technical field of water treatment agent preparation. The modified nano titanium dioxide is obtained by coating titanium dioxide with 3-methyl acrylic acid sulfopropyl potassium salt grafted polydopamine. The titanium dioxide has photocatalytic effect and antibacterial property, can catalytically decompose pollutants in sewage, inhibit the breeding of bacteria in water, and can catalytically decompose free formaldehyde in dicyandiamide formaldehyde polycondensate. The modified nano titanium dioxide is loaded on phenolic resin foam, and after calcination, the nano titanium dioxide / foam carbon powder is formed. The large specific surface area and porous property of the foam carbon can increase the capture and fixation of free formaldehyde, and can synergistically accelerate the catalytic degradation of formaldehyde with the titanium dioxide. The dicyandiamide formaldehyde polycondensate grafted with castor oil is in-situ polymerized on the surface of the matrix. The castor oil grafting can significantly improve the dicyandiamide formaldehyde polycondensate, significantly increase the molecular weight, and increase the capture capacity.

A multi-stage adsorption synergistic dust control system for silicon carbide laser processing includes a suction hood, a vibrating grid module, a porous laminar flow adsorption module, a boiling water mist module, a condensation pipe, a mixing condensation module, a drying module, and a fan, arranged sequentially and connected along the airflow direction. The vibrating grid module includes an alloy grid and a vibration device. The porous laminar flow adsorption module includes an adsorption layer disposed in the airflow channel. The boiling water mist module includes an adsorption chamber and a water mist generator to form boiling water mist to capture dust in the airflow. The condensation pipe performs preliminary condensation of the airflow to capture dust. The mixing condensation module includes an inclined condensation channel that utilizes dust in the airflow as condensation nuclei to remove dust. The inner wall of the condensation channel is alternately arranged with superhydrophilic and superhydrophobic regions. This invention can achieve stepwise dust removal from the micron to the nanometer scale, with a capture efficiency of up to 99.9%, low overall energy consumption, and recyclable materials.

The application relates to the technical field of vehicle exhaust treatment, and provides an exhaust treatment system and a vehicle. The exhaust treatment system comprises a particle trap and a regeneration device, the regeneration device comprises an oil-gas mixing part, a gas path assembly and an oil path assembly, and a fluid driving part connected between the oil-gas mixing part and the particle trap, and the fluid driving part can drive an oil-gas mixture in the oil-gas mixing part to flow into the particle trap. The exhaust treatment system provided by the application sets one fluid driving part, which can facilitate the smooth flow of exhaust gas and fuel into the oil-gas mixing part, the exhaust treatment system has a simple structure, and the oil-gas mixture can smoothly flow to the downstream, which is helpful for the combustion regeneration of the particle trap, has a good exhaust treatment effect, and can reduce the emission of pollutants.

This invention relates to the field of crop identification technology, specifically to a UAV image recognition system for crops in karst landforms based on multi-scale features. The system includes: a feature database construction module for constructing a hierarchical crop feature database, wherein the crop feature database comprises multiple storage areas storing feature data of different crop types, each storage area containing multiple feature layers, and multiple feature layers located in the same storage area are associated with each other; an image acquisition and preprocessing module for acquiring remote sensing images of the area to be identified collected by the UAV, and performing geometric correction, spectral correction, and standardized slice preprocessing on the remote sensing images to obtain the image to be analyzed; and a feature extraction and matching module for communicating with the feature database construction module, thereby improving the efficiency and accuracy of crop type identification.

This invention relates to a flame retardant with both microwave absorption and smoke suppression functions. The flame retardant is a hierarchical porous metal-organic framework material with a three-dimensional polygonal structure, a size of 1-3 μm, and a pore size distribution between 1.0-50.0 nm. It carries nano-transition metal oxides internally and on its surface, with the nano-transition metal oxides accounting for 25%-65% of the total mass. The flame retardant provided by this invention possesses microwave absorption, flame retardancy, and smoke suppression properties. Its initial decomposition temperature is 442℃±10℃, with significant mass loss in the temperature range of 450-700℃. At 800℃, the residue content is as high as 50% or more, exhibiting excellent thermal stability and meeting the processing requirements for flame retardants in polyurea, epoxy resin, and polycarbonate.

The utility model discloses an efficient intelligent phosphorite flotation cell which comprises a flotation cell body, a plurality of partition plates are arranged in the flotation cell body to divide the flotation cell body into a plurality of cell bodies, a stirring device is arranged in each cell body, grooves for foam to flow out are formed in the top ends of the two sides of the flotation cell body, and a laser foam breaker is arranged at one end of each groove. The aeration holes of the air distributor are improved, bubbles are generated more sufficiently, capturing of hydrophobic mineral particles is facilitated, the flotation effect is better, and the flotation efficiency is higher. An infrared liquid level instrument installed on a phosphorite flotation tank body is in linkage with an air distributor installed at the bottom end of a flotation machine stirring paddle, the height of a foam layer is detected through infrared liquid level scanning, when the foam layer is too high, the air distributor automatically reduces the inflation amount and reduces generation of bubbles, and when the foam layer is too low, the air distributor automatically increases the inflation amount and increases generation of bubbles. And intelligentization and automation of the equipment are realized. And the laser bubbles are adopted, so that the bubbles can be broken more quickly and thoroughly, and the water consumption is also reduced.

The invention discloses a beetle elytra-imitated interlocking structure, a metamaterial plate comprising the beetle elytra-imitated interlocking structure and a composite armor system, and belongs to the field of multifunctional advanced protective materials. The interlocking structure comprises a plurality of functional elements arranged in a planar array, the adjacent functional elements are mechanically connected through geometric interlocking interfaces, the contact surfaces of the geometric interlocking interfaces are in a non-chemical bonding state and can generate relative dislocation under an impact load, and the geometric interlocking interfaces are in a non-chemical bonding state. Each functional element is integrally prepared from a polymer-based composite material dispersed with an electromagnetic loss filler, the interior of each functional element is of a three-dimensional porous micro-lattice structure with a negative Poisson's ratio effect, and the porosity and / or cell size of each three-dimensional porous micro-lattice structure are / is changed in a gradient manner in the thickness direction of the corresponding functional element. The metamaterial plate and the composite armor system using the interlocking structure can effectively inhibit structural layering under an impact load, can actively improve anti-penetration capability, and also have an excellent broadband electromagnetic wave absorption function.

The application provides a kitchen waste sorting device, and belongs to the technical field of waste treatment. The kitchen waste sorting device comprises a material distribution mechanism, a plurality of plastic cutting members and a double-curved-surface stirrer, and a conical material distribution hopper is provided with an inlet pipe on the side. An air flotation tank is located above the conical material distribution hopper and is in communication with the top of the conical material distribution hopper. The sidewall of the air flotation tank is provided with a discharge pipe. A plastic conveying mechanism is located below the water surface of the air flotation tank at the bottom and is located obliquely above the air flotation tank at the top. A slag discharge mechanism comprises a discharge pipe and a screw conveyor. The integrated kitchen waste sorting device can realize efficient sorting of plastic, waste oil, food residue and waste water. The waste oil and food residue can be recycled, which is conducive to the recycling of kitchen waste and the subsequent solid waste and waste water treatment.

The invention discloses a 3D speaking face generation method of emotion controllable VQ-VAE based on hierarchical decoupling. The method specifically comprises the steps of 1, preprocessing a data set; 2, constructing a first-stage model, setting a loss function and training the model; 3, constructing a second-stage model, setting a loss function and training the model; and step 4, constructing a generative model, and generating a 3D speaking face video. According to the method, through the condition decoupling based on VQ-VAE, facial expressions and actions generated by the 3D speaking face generation model based on condition guidance are more accurate and real, and accurate and stable 3D speaking face generation is realized.

This invention belongs to the field of photocatalyst materials technology, specifically relating to a three-dimensional graphene aerogel photo-self-propelled Fenton catalyst, its preparation method, and its application. The preparation method includes: mixing a graphene dispersion containing ethylenediamine with a phenolic resin prepolymer containing ferrocene, phenols, and an alkaline catalyst; followed by a one-step hydrothermal reaction, washing, and freeze-drying to obtain the three-dimensional graphene aerogel photo-self-propelled Fenton catalyst. This catalyst can be used to degrade organic pollutants in water under visible light irradiation.

This invention belongs to the field of gas-liquid separation technology, and relates to a gas-liquid separator for hydrogen production by water electrolysis. It includes a separation tank and a gas-liquid separation component disposed within the separation tank. The gas-liquid separation component includes a support column and a spiral guide baffle sleeved on the support column. The support column is rotatably connected to the separation tank. The spiral guide baffle includes a base column and multiple spiral curved blades disposed on the side wall of the base column, with the multiple spiral curved blades evenly distributed circumferentially on the base column. The base column is sleeved on the support column and rotatably connected to it. A gas outlet, a gas-liquid mixture inlet, and a liquid outlet are sequentially arranged from top to bottom on the side wall of the separation tank. The gas-liquid mixture inlet is directly opposite the spiral guide baffle. This invention can ensure dynamic and stable gas-liquid separation under different operating conditions and is suitable for efficient gas-liquid separation in hydrogen production by water electrolysis, where gas production is extremely high.

The application provides a kind of pedestrian trajectory prediction method based on double-branch space-time graph convolution network fusion, belongs to the field of mobile robot perception and decision-making, solve the technical problems that pedestrian social interaction is insufficient in traditional method, and pedestrian movement mode is difficult to capture, need to simulate complex time dependence, its technical scheme is: including the following steps: step one, using pre-training CNN extracts pedestrian scene features;Step two, in double branch, respectively construct pedestrian two two interaction graph and group interaction graph, and embed scene features in the graph;Step three, using spatial GCN extracts pedestrian spatial features;Step four, using time extrapolation convolution network processes the space-time features output by two branches to predict the Gaussian distribution parameters of trajectory.The beneficial effects of the application are that the application can better model the complex social interaction of pedestrians and capture the unique movement patterns of each pedestrian, improving the accuracy and generalization ability of the prediction.

The application provides a kind of carbon nanoparticles-based tidal power generation device, and relates to the technical field of tidal power generation, comprising: floating foundation platform and power generation unit, the floating foundation platform includes base, buoyancy bucket, support rod, power generation unit slot and rotation control system;The base is hollow cuboid, the base is internally provided with rotation control system, the rotation control system is connected with several support rods, a plurality of power generation unit slots are provided on each support rod, the base is connected with buoyancy bucket on both sides through clamping groove, and the base rear side is provided with locking mechanism;The power generation unit is arranged in the power generation unit slot, and the power generation units on each support rod are connected in series with the power generation units on another row in series or parallel connection after being connected in series with each other.The application has the characteristics of low cost and simple preparation, and the formed tidal power generation device can realize stable power output.

The present application belongs to the field of direct current transmission, and particularly relates to a support insulator, an insulating support structure and a direct current GIL. In order to solve the technical problem that the electric field lines pass through the support insulator in the prior art, resulting in that the normal field intensity of the side surface of the support insulator is too large, the present application provides a support insulator, which comprises a high-potential support surface for supporting a conductive rod, a low-potential support surface for supporting an insulating cylinder and a side surface, the high-potential support surface and the low-potential support surface are both arc surfaces, the axis of the high-potential support surface and the low-potential support surface corresponds to the axis of the matched conductive rod, and the extension line of the projection of the side surface of the support insulator in the direction of the axis of the corresponding conductive rod intersects with the axis of the conductive rod. The present application further provides an insulating support structure and a direct current GIL using the support insulator. By making the electric field lines parallel to the side surface of the support insulator, the normal field intensity of the side surface of the support insulator can be reduced, and the surface flashover voltage and the insulating strength of the support insulator can be improved.

This invention belongs to the field of blast furnace carbon emission reduction technology and discloses a blast furnace carbon emission reduction ironmaking system and method, including: oxygen-enriched ironmaking with the participation of a mixed reaction gas with high CO2 concentration. Due to the use of CO2 in the blast furnace blast and pulverized coal carrier gas, the oxygen enrichment rate is increased, and the amount of biochar is increased, thereby improving the smelting intensity of the blast furnace, increasing the proportion of pulverized coal replacing coke (increasing the coal ratio and decreasing the coke ratio), reducing ironmaking costs, and reducing carbon emissions. Furthermore, due to the use of CO2 injection and oxygen-enriched ironmaking operations, the amount of air incorporated into the mixed reaction gas is reduced, increasing the concentration of CO and CO2 in the blast furnace gas, which is beneficial for subsequent capture, separation, purification, and resource utilization of different components in the blast furnace gas.

The utility model relates to the technical field of optical detection, in particular to an AOI detection light source and a detection system. The detection light source comprises a first light source, a second light source and a third light source; part of light emitted by the first light source is condensed by the cylindrical mirror and obliquely reflected by the semi-reflecting and semi-transmitting mirror to form a first light beam; the two groups of second light sources are symmetrically arranged, and light emitted by each group of second light sources is obliquely reflected by the first light guide surface and the second light guide surface in sequence and then converged to form a second light beam; the two third light sources are symmetrically arranged, light emitted by each third light source is obliquely reflected by the third light guide surface and then converged to form a third light beam, and the third light beam is located on the side, away from the first light beam, of the second light beam; the total angle of the first light beam, the second light beam and the third light beam is larger than or equal to 160 degrees. According to the AOI detection light source disclosed by the utility model, due to the unique optical design, light rays up to 160 degrees can be incident to a defect area in a convergent form, so that a camera can conveniently capture defect characteristic information.

The application relates to the technical field of hydrogen pipeline safety experiment testing and accident consequence measurement, and discloses a hydrogen pipeline jet fire heat radiation sensor arrangement method and system, which comprises the following steps: a three-dimensional geometric model identical to a high-pressure hydrogen jet fire experiment site is constructed; test working conditions and site constraints are input into the three-dimensional geometric model, a numerical simulation scene consistent with target working conditions is constructed; for each working condition, heat radiation distribution maps of multiple height layers are simulated and output; according to the heat radiation distribution maps, candidate measuring points are divided, and predicted heat flux values at each candidate measuring point are extracted; according to the positions of the candidate measuring points, a heat radiation sensor arrangement optimization model is established, and an optimal sensor arrangement scheme is determined; according to the sensor arrangement scheme, a high-pressure hydrogen jet fire experiment is carried out, and the heat radiation sensor arrangement optimization model is optimized. By using the application scheme, high-reliable capture of jet fire heat radiation distribution characteristics can be realized.

The application discloses a power prediction method and device, equipment and storage medium, and relates to the technical field of new energy. The method comprises the following steps: acquiring an original historical power time sequence, and decomposing the original historical power time sequence to obtain target power time sequences in a plurality of different frequency band ranges; an improved xLSTM model is used to output initial power prediction sequences of the plurality of different frequency bands according to the target power time sequences, wherein information fusion weights in the improved xLSTM model are determined according to fluctuations of power in the original historical power time sequence; and the initial power prediction sequences of the plurality of different frequency bands are mixed to obtain target power prediction. The technical scheme of the embodiment of the application has higher prediction accuracy and lower prediction error.